Isabelle Guyon (born August 15, 1961, in Paris, France) is a French, Swiss, and American computer scientist and researcher in machine learning and artificial intelligence.^[1,35,2] She studied at the École supérieure de physique et de chimie industrielles de la ville de Paris (ESPCI Paris), graduating with a master's degree in 1985, and completed her doctorate in 1988 under Gérard Dreyfus at Pierre and Marie Curie University (University of Paris VI) with a dissertation on neural networks in pattern recognition (Réseaux de neurones pour la reconnaissance des formes : architectures et apprentissage).

Afterward, she was a postdoctoral researcher and group leader at Bell Laboratories, where she worked on handwriting recognition and Siamese neural networks. In 1996, she left Bell Laboratories and moved to Berkeley, California, where she raised her three children and founded the machine learning company Clopinet, applying machine learning to identify genes involved in cancer. In the 2000s, Guyon organized machine learning competitions to make them generally accessible, founding the organization ChaLearn in 2011 in Bear Valley, California. In 2017, she presided over the Conference on Neural Information Processing Systems (NeurIPS).

Guyon became a full professor at Université Paris-Saclay in 2016, holding the Big Data chair, while also conducting research at INRIA and participating in the TAU (TAckling the Underspecified) group at the University of Paris-South.^[2,5,1] Since October 2022, she has also served as a Director of Research at Google DeepMind.^[3,5,1] Considered a pioneer in machine learning following her 1992 co-development of support-vector machines (SVM) alongside Vladimir Vapnik, Bernhard Boser, and Bernhard Schölkopf, Guyon focuses her research on artificial neural networks, feature subset selection, and bioinformatics.^[1,4,5,2] Among the highly cited scientists in her field, she was named an AMIA Fellow by the American Medical Informatics Association in 2011, received the BBVA Foundation Frontiers of Knowledge Award with Bernhard Schölkopf and Vladimir Vapnik in 2019 according to German Wikipedia or in 2020 according to English and Hebrew Wikipedia, and was named Chevalier de la Légion d'Honneur in 2026.

Isabelle Guyon was born on August 15, 1961 in Paris. After graduating from the French engineering school ESPCI Paris in 1985, she joined Gérard Dreyfus's laboratory at Pierre and Marie Curie University to complete her doctorate on neural network architectures and training.^[7,8,6,32] Guyon defended her thesis in 1988 and was hired the following year at AT&T Bell Laboratories, first as a postdoctoral researcher and later as a group leader.^[10] She worked at Bell Labs for six years, exploring research fields ranging from neural networks to pattern recognition algorithms and computational learning theory, with handwriting recognition as a primary application.^[9] During her time there, she collaborated with Yann LeCun, Léon Bottou, Vladimir Vapnik, Corinna Cortes, Yoshua Bengio, and Patrice Simard, and met her future husband, Bernhard Boser.^[1]

In 1996, Guyon left Bell Laboratories and raised her children in Berkeley, California, where she founded her own machine learning consulting company, Clopinet.^[10,11,22] She took an interest in medical applications, using her previous work to classify the genes responsible for various types of cancer.^[11,12,38] Since 2003, Guyon has organized numerous data science challenges and competitions to stimulate research in the field.^[33] In 2011, she founded ChaLearn, a non-profit organization dedicated to creating and running machine learning challenges open to everyone, serving as its founding president.^[12,13,14] Since 2015, she has also served as the Community Lead for the CodaLab Competitions platform on behalf of Paris-Saclay University.^[11]

In late 2013, Guyon undertook a five-month residency in France at Aix-Marseille University within the Qarma team of the Laboratoire d'Informatique Fondamentale de Marseille, which enabled her to establish local, national, and European collaborations. In 2016, she returned to France to take up the Big Data Chair professorship held jointly between Paris-Saclay University and INRIA, and works with the TAU (TAckling the Underspecified) group at the Laboratoire de recherche en informatique.^[4,18,12] Guyon served as Program Chair of NeurIPS 2016 and General Chair of NeurIPS in 2017, is an associate and action editor for the Journal of Machine Learning Research, serves as series editor for Series: Challenges in Machine Learning, and is a member of the European Laboratory for Learning and Intelligent Systems.^{[14,17,15,16,34,30]} Together with Bernhard Schölkopf and Vladimir Vapnik, she received the BBVA Foundation Frontiers of Knowledge Award in 2020 for her work in machine learning and artificial intelligence. Since October 2022, she has been a director and research scientist at Google DeepMind labs, and in 2024, she was elected as a member of the French Academy of Technologies.^[1,18]

Among her most notable contributions, Guyon co-invented support-vector machines (SVM) in 1992 with Bernhard Boser and Vladimir Vapnik.^[19] SVM is a supervised machine learning algorithm, comparable to artificial neural networks or decision tree learning algorithms, that quickly became a classical technique in machine learning. SVMs have especially contributed to the popularization of kernel methods.

During her years at Bell Laboratories, Isabelle Guyon took part in numerous projects involving neural networks. In particular, she co-authored some of the first papers on the use of neural networks for handwriting recognition to recognize handwritten digits using the MNIST database.^[20,36] She is also a co-inventor of Siamese neural networks, a neural network architecture and similarity learning algorithm used to learn similarities, with applications to signature, face, or object recognition.^[11]

Isabelle Guyon is the author of numerous publications at the intersection of biology (cancer research and genomics) and artificial intelligence. She notably introduced the use of support-vector machines to detect cancer and determine its presence in an individual using genes.^[21,37]

Through her non-profit organization ChaLearn, Isabelle Guyon organized and directed challenges open to everyone in order to solve open problems in machine learning across diverse fields, including computer vision, neuroscience, particle physics, feature selection, causality, and automated machine learning.^[27,22,24] Most of the challenges organized by ChaLearn have resulted in publications, among the most cited of which are:^{[39,40,41,42]} Guyon et al., Result analysis of the NIPS 2003 feature selection challenge, Advances in neural information processing systems, 2005, link; Escalera et al., ChaLearn Looking at People Challenge 2014: Dataset and Results, Computer Vision - ECCV 2014 Workshops, Springer International Publishing, 2014, link; Guyon et al., A brief Review of the ChaLearn AutoML Challenge, JMLR: Workshop and Conference Proceedings 64:21-30, 2016, link; and Adam-Bourdario et al., The Higgs boson machine learning challenge, JMLR: Workshop and Conference Proceedings 42:19-55, 2015, link.

Isabelle Guyon is married to Bernhard Boser, a professor at UC Berkeley.^[28] She has twin boys and a daughter, all three of whom have pursued scientific studies and completed a science degree.^[29] She holds three citizenships: French by birth, Swiss by marriage, and American by naturalization.^[30,1]

She was named an American Medical Informatics Association Fellow in 2011, received the BBVA Foundation Frontiers of Knowledge Award in 2020, and was nominated to the French Academy of Technologies in 2024. Additionally, she is a member of the European Laboratory for Learning and Intelligent Systems.

In 1992, Bernhard Boser, Isabelle Guyon, and Vladmir Vapnik published "A training algorithm for optimal margin classifiers" in the Proceedings of the fifth annual workshop on Computational learning theory (doi:10.1145/130385.130401). Guyon co-authored "Signature verification using a 'siamese' time delay neural network" alongside Jane Bromley, Yann LeCun, Eduard Sckinger, and Roopak Shah, which appeared in Advances in Neural Information Processing Systems in 1994. In 2002, Guyon, Jason Weston, Stephen Barnhill, and Vladimir Vapnik published "Gene selection for cancer classification using support vector machines" in Machine Learning, published by Kluwer Academic Publishers (doi:10.1023/A:1012487302797). In 2003, Guyon and Andr Elisseeff published "An introduction to variable and feature selection" in the Journal of Machine Learning Research.